github/yt-dlp
1
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2024-11-07 20:30:41 -05:00
Code Issues Projects Releases Packages Wiki Activity

[youporn] Fix title and description extraction (closes #18748)

Browse source
This commit is contained in:
Sergey M․ 2019-01-09 00:37:01 +07:00
parent 2543938bbe
commit 6089ff40e7
No known key found for this signature in database
GPG key ID: 2C393E0F18A9236D
1 changed files with 8 additions and 6 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

14
youtube_dl/extractor/youporn.py
View file

@ -68,11 +68,9 @@ def _real_extract(self, url):
request.add_header('Cookie', 'age_verified=1')
webpage = self._download_webpage(request, display_id)
title = self._search_regex(
[r'(?:video_titles|videoTitle)\s*[:=]\s*(["\'])(?P<title>(?:(?!\1).)+)\1',
r'<h1[^>]+class=["\']heading\d?["\'][^>]*>(?P<title>[^<]+)<'],
webpage, 'title', group='title',
default=None) or self._og_search_title(
title = self._html_search_regex(
r'(?s)<div[^>]+class=["\']watchVideoTitle[^>]+>(.+?)</div>',
webpage, 'title', default=None) or self._og_search_title(
webpage, default=None) or self._html_search_meta(
'title', webpage, fatal=True)
@ -134,7 +132,11 @@ def _real_extract(self, url):
formats.append(f)
self._sort_formats(formats)
description = self._og_search_description(webpage, default=None)
description = self._html_search_regex(
r'(?s)<div[^>]+\bid=["\']description["\'][^>]*>(.+?)</div>',
webpage, 'description',
default=None) or self._og_search_description(
webpage, default=None)
thumbnail = self._search_regex(
r'(?:imageurl\s*=|poster\s*:)\s*(["\'])(?P<thumbnail>.+?)\1',
webpage, 'thumbnail', fatal=False, group='thumbnail')